4 datasets found

The dataset used in the paper is a personalized news recommendation dataset. The goal is to learn to recommend articles to any user (depending on their contextual features) to...

The dataset is a two-mode tensor depicting binary (user, anime) preferences. The tensor contains 1, 300, 160 observed entries, of size 25, 838 × 4, 066.

A simplified simulated dataset showing varying user preferences over time and how this can be evaluated with static and dynamic learning algorithms.

A simulated dataset of diverse, high-level preferences representing a population, used to pretrain good initializations and adapt to user preferences.